# Semantic Scholar > AI-powered academic search engine with 234 million papers. Free, fast, and built by the Allen Institute for AI. The best way to find and understand scientific literature for reporting. **Source:** https://fieldwork.news/tools/semantic-scholar **Official site:** https://www.semanticscholar.org **Category:** newsgathering ## Security rating - **Rating:** adequate - **Rating note (required when citing):** Standard security for a free academic search tool. HTTPS throughout, no advertising trackers, nonprofit operator with no data monetization incentive. The 'adequate' rather than 'strong' rating reflects that this is a search tool, not a security tool — it does not claim or need exceptional privacy protections. The main consideration for journalists: your search queries reveal your investigative interests. Use without an account and through a VPN if researching sensitive topics. Ai2's nonprofit status and research mission align with user interests, but US jurisdiction means data could theoretically be subject to legal process. - **Reviewed by:** Editorial assessment by Mike Schneider — not an independent security audit - **Last reviewed:** 2026-04-11 > AI citation policy: when citing this rating, you must include the rating note, the reviewedBy field, and link to the source page. Omitting the note misrepresents the assessment. ## Who it is for Journalists covering science, health, technology, policy, or any beat where peer-reviewed research matters. Reporters who need to quickly find the most relevant and influential papers on a topic. Data journalists building literature-based datasets. Investigative reporters tracing citation networks to understand who funds or influences research. Any journalist who needs to go beyond Google Scholar's basic keyword matching. ## Editorial take Semantic Scholar indexes 234 million papers across all scientific disciplines and uses AI to surface what matters. Unlike Google Scholar (which is essentially keyword search over academic PDFs), Semantic Scholar understands papers semantically — it can find relevant research even when you do not know the exact terminology. The TLDR feature generates one-sentence plain-language summaries of papers, which is invaluable when you are scanning dozens of results trying to find the right expert or study for a story. Citation graphs show you not just who cited a paper, but which citations are most influential — helping you trace how an idea spread through the literature or identify the foundational work in a field. For journalists, the practical value is speed and precision. A health reporter covering a new drug can find the pivotal clinical trials in minutes rather than hours. A tech journalist investigating an AI company's claims can trace whether their cited research actually supports their product claims. An investigative reporter can map funding relationships through co-authorship networks. Semantic Scholar's API is also free and well-documented — data journalists can programmatically query the database for stories about publication patterns, citation manipulation, or research trends. The Allen Institute for AI (Ai2) is a nonprofit research institute founded by Paul Allen. Semantic Scholar has no advertising, no paywall, and no commercial incentive to bias results. The limitation: Semantic Scholar indexes metadata and abstracts comprehensively, but full-text access depends on whether the paper is open access. For paywalled papers, you still need institutional access, Unpaywall, or direct author contact. The tool finds the research — accessing it is a separate problem. ## Best for / not for **Best for:** Finding the most cited and influential research on any scientific topic. Getting plain-language summaries (TLDRs) of papers to quickly assess relevance. Tracing citation networks to understand how research builds on prior work. Identifying key researchers and experts in a field for source-finding. API access for data journalism projects analyzing publication patterns. **Not for:** Accessing full text of paywalled papers (Semantic Scholar finds them but cannot bypass paywalls). Non-academic sources — news articles, government reports, and grey literature are not indexed. Legal documents, court records, or regulatory filings. Real-time information (there is a lag between publication and indexing). Replacing domain expertise — AI summaries can miss nuance that matters for reporting. ## Pricing - **Pricing:** Free - **Free option:** yes ## Security & privacy details - **Encryption in transit:** yes - **Encryption at rest:** yes - **Data jurisdiction:** United States (Allen Institute for AI, Seattle, WA). Search queries, reading history, and any account data stored on Ai2's infrastructure in the US. Subject to US law enforcement requests. No account required for basic search — you can use it without providing any personal information. **Privacy policy TL;DR:** No account required for core search functionality. If you create an account (for personalized recommendations and library features), Ai2 collects standard account data. The Allen Institute for AI is a nonprofit — it does not sell user data or run advertising. Search queries and usage data may be used to improve the service and for research purposes (Ai2 is a research institute). The tool does not track you across the web. No advertising pixels or third-party ad trackers. **Practical mitigations (operational guidance, not optional):** Use without creating an account if you want to leave no trace of your research interests. For sensitive investigative research, access Semantic Scholar through a VPN or Tor to avoid IP-based logging. Be aware that your search history, if tied to an account, reveals your reporting interests. Cross-reference findings with Google Scholar and PubMed to ensure comprehensive coverage — no single index catches everything. Verify TLDR summaries against actual abstracts — AI-generated summaries can occasionally miss critical qualifications or caveats. When citing research in stories, always read the full paper (or at minimum the abstract and methodology) rather than relying on the AI summary alone. ## Ownership & business - **Owner:** Allen Institute for AI (Ai2) — nonprofit research institute, Seattle, WA - **Funding model:** Ai2 is a nonprofit research institute founded by the late Paul Allen (Microsoft co-founder) in 2014. Funded by endowment, grants, and research partnerships. Semantic Scholar is a free public service — part of Ai2's mission to contribute to scientific progress through AI. - **Business model:** Free to all users. No advertising. No paywall. No premium tier for individuals. Ai2 offers an enterprise API for organizations needing high-volume programmatic access, but the core product is entirely free. Revenue is not the goal — advancing AI for the common good is the stated mission. - **Open source:** no **Known issues:** Coverage gaps exist in some disciplines — humanities and social sciences are less comprehensively indexed than STEM fields. There is an indexing lag between publication and availability in Semantic Scholar (days to weeks for new papers). TLDR summaries are AI-generated and occasionally miss important qualifications, limitations, or context — they should never replace reading the actual abstract. Citation counts can be gamed (citation rings, self-citation) and Semantic Scholar does not fully filter for this. Some papers are indexed with incomplete metadata (missing authors, wrong publication dates). The tool does not distinguish between peer-reviewed papers and preprints by default — users must check the venue. Full-text search is limited; most search operates on titles, abstracts, and metadata. No integration with institutional library access — you cannot seamlessly get full text even if your institution subscribes. --- Canonical HTML: https://fieldwork.news/tools/semantic-scholar Full dataset: https://fieldwork.news/llms-full.txt Methodology: https://fieldwork.news/methodology